Runtime feedback in a meta-tracing JIT for efficient dynamic languages

Bolz, Carl Friedrich; Cuni, Antonio; Fijałkowski, Maciej; Leuschel, Michaël; Pedroni, Samuele; Rigo, Armin

doi:10.1145/2069172.2069181

Cited by 36 publications

(26 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By annotating their interpreters with certain hints [5], e.g., for detection of loops in the user-program, language developers can guide tracing and optimization of traces. The traces are then heavily optimized to produce efficient machine code corresponding to the relevant operations performed by the user program.…”

Section: Difference With Meta-tracingmentioning

confidence: 99%

A flexible framework for studying trace-based just-in-time compilation

Vandercammen

Marr

Roover

2018

Computer Languages, Systems & Structures

View full text Add to dashboard Cite

Just-in-time compilation has proven an effective, though effort-intensive, choice for realizing performant language runtimes. Recently introduced JIT compilation frameworks advocate applying meta-compilation techniques such as partial evaluation or meta-tracing on simple interpreters to reduce the implementation effort.However, such frameworks are few and far between. Designed and highly optimized for performance, they are difficult to experiment with. We therefore present STRAF, a minimalistic yet flexible Scala framework for studying trace-based JIT compilation. STRAF is sufficiently general to support a diverse set of language interpreters, but also sufficiently extensible to enable experiments with trace recording and optimization. We demonstrate the former by plugging two different interpreters into STRAF. We demonstrate the latter by extending STRAF with e.g., constant folding and type-specialization optimizations, which are commonly found in dedicated trace-based JIT compilers. The evaluation shows that STRAF is suitable for prototyping new techniques and formalisms in the domain of trace-based JIT compilation.

show abstract

Section: Difference With Meta-tracingmentioning

confidence: 99%

A flexible framework for studying trace-based just-in-time compilation

Vandercammen

Marr

Roover

2018

Computer Languages, Systems & Structures

View full text Add to dashboard Cite

show abstract

“…Both PyPy and Pyrolog are optimised for meta-tracing, in the sense that their implementation has, where necessary, been structured according to meta-tracings demands. Such structuring is relatively minor (see [7] for more details): most commonly, tracing annotations [4] are added to the interpreter to provide hints (e.g. "this RPython function's loops can safely be unrolled") and guarantees (e.g.…”

Section: Interaction With Meta-tracingmentioning

confidence: 99%

“…A tracing JIT's natural tendency to aggressively typespecialise code (see [4,17]) is important in reducing the overhead of object conversions between interpreters. Tracing a call from one language to the other naturally traces the object conversion code (Section 5.3), type specialising it.…”

Section: Interaction With Meta-tracingmentioning

confidence: 99%

Unipycation

Barrett

Bolz

Tratt

2013

Proceedings of the 7th ACM Workshop on Virtual Machines and Intermediate Languages

Self Cite

View full text Add to dashboard Cite

Language composition approaches have traditionally suffered from poor performance. In this paper we hypothesise that meta-tracing provides a means to compose independent language interpreters while retaining the performance levels of each. To study this approach, we compose Python and Prolog interpreters to form Unipycation. We present a case study of its use and a suite of micro-benchmarks which give us some understanding of its cross-language performance.

show abstract

“…Previous implementations have used either static compilers, e.g., Renz's solver [2] and GQR [3], or method-based JIT compilers, e.g., the RCC-8 reasoning module of the description logic reasoner PelletSpatial [4]. The advantage of trace-based JIT compilers is that they can discover optimization opportunities in common dynamic execution paths, that are not apparent to a static compiler or a method-based JIT compiler [10], [11].…”

Section: Pyrcc8 -A State Of the Art Qsrmentioning

confidence: 99%

“…• We make the case for a new generation of RCC-8 reasoners implemented in Python, a general-purpose, interpreted high-level programming language which enables rapid application development, and making use of advanced Python environments, such as PyPy 2 , utilizing trace-based just-in-time (JIT) compilation techniques [10], [11]. We present such a reasoner, called PyRCC8, and compare it to other well-known reasoners from the literature [2]- [4].…”

Section: Introductionmentioning

confidence: 99%

Consistency of Chordal RCC-8 Networks

Sioutis

Koubarakis

2012

2012 IEEE 24th International Conference on Tools With Artificial Intelligence

View full text Add to dashboard Cite

Abstract-We consider chordal RCC-8 networks and show that we can check their consistency by enforcing partial path consistency with weak composition. We prove this by using the fact that RCC-8 networks with relations from the maximal tractable subsetsĤ8, C8, and Q8 of RCC-8 have the patchwork property. The use of partial path consistency has important practical consequences that we demonstrate with the implementation of the new reasoner PyRCC8 , which is developed by extending the state of the art reasoner PyRCC8. Given an RCC-8 network with only tractable RCC-8 relations, we show that it can be solved very efficiently with PyRCC8 by making its underlying constraint graph chordal and running path consistency on this sparse graph instead of the completion of the given network. In the same way, partial path consistency can be used as the consistency checking step in backtracking algorithms for networks with arbitrary RCC-8 relations resulting in very improved pruning for sparse networks while incurring a penalty for dense networks.

show abstract

Runtime feedback in a meta-tracing JIT for efficient dynamic languages

Cited by 36 publications

References 19 publications

A flexible framework for studying trace-based just-in-time compilation

A flexible framework for studying trace-based just-in-time compilation

Unipycation

Consistency of Chordal RCC-8 Networks

Contact Info

Product

Resources

About